Patent Application
20190394083
The present disclosure relates generally to time-series data forecasting. More specifically, but not by way of limitation, this disclosure relates to a pipeline system for time-series data forecasting using a distributed computing environment.
A time series is a series of data points indexed in time order, such as a series of data collected sequentially at a fixed time interval. Time-series forecasting is the use of a model to predict future values for a time series based on previously observed values of the time series.
One example of the present disclosure includes a system. The system can include a processing device and a memory device comprising instructions that are executable by the processing device. The instructions can cause the processing device to access a pipeline for forecasting a plurality of time series. The pipeline represents a sequence of operations for processing the plurality of time series to produce forecasts. The sequence of operations includes model strategy operations for applying model strategies to the plurality of time series to determine error distributions corresponding to the model strategies. Each of the model strategies is configured to receive the plurality of time series and determine champion forecasts for the plurality of time series. The champion forecasts can be determined by, for each time series among the plurality of time series: generating respective forecasts for the time series by applying forecasting models to the time series, and determining a respective champion forecast for the time series by comparing error metrics associated with the respective forecasts. Each of the model strategies is further configured to determine error values associated with the champion forecasts and generate one of the error distributions using the error values. Each of the error values indicates the accuracy of a respective champion forecast among the champion forecasts. The sequence of operations also includes a model-strategy comparison operation, which is configured to receive the error distributions corresponding to the model strategies and determine aggregate error values associated with the model strategies based on the error distributions. Each aggregate error value is associated with one of the model strategies and is determined by aggregating the error values in one of the error distributions. The model-strategy comparison operation is further configured to determine which of the model strategies is a champion model strategy for the plurality of time series by comparing the aggregate error values to one another. The instructions can further cause the processing device to execute the pipeline to determine the champion model strategy for the plurality of time series.
Another example of the present disclosure includes a method. The method includes accessing a pipeline for forecasting a plurality of time series, the pipeline representing a sequence of operations for processing the plurality of time series to produce forecasts. The sequence of operations includes model strategy operations for applying model strategies to the plurality of time series to determine error distributions corresponding to the model strategies. Each of the model strategies is configured to receive the plurality of time series and determine champion forecasts for the plurality of time series by, for each time series among the plurality of time series, generating respective forecasts for the time series by applying forecasting models to the time series and determining a respective champion forecast for the time series by comparing error metrics associated with the respective forecasts. Each of the model strategies is further configured to determine error values associated with the champion forecasts, each of the error values indicating the accuracy of a respective champion forecast among the champion forecasts and generate one of the error distributions using the error values. The sequence of operations also includes a model-strategy comparison operation, which is configured to receive the error distributions corresponding to the model strategies and determine aggregate error values associated with the model strategies based on the error distributions. Each aggregate error value is associated with one of the model strategies and is determined by aggregating the error values in one of the error distributions. The model-strategy comparison operation is configured to determine which of the model strategies is a champion model strategy for the plurality of time series by comparing the aggregate error values to one another. The method further includes executing the pipeline to determine the champion model strategy for the plurality of time series.
Yet another example of the present disclosure includes a non-transitory computer readable medium comprising instructions that are executable by a processing device. The instructions can cause the processing device to access a pipeline for forecasting a plurality of time series. The pipeline represents a sequence of operations for processing the plurality of time series to produce forecasts. The sequence of operations includes model strategy operations for applying model strategies to the plurality of time series to determine error distributions corresponding to the model strategies. Each of the model strategies is configured to receive the plurality of time series and determine champion forecasts for the plurality of time series by, for each time series among the plurality of time series, generating respective forecasts for the time series by applying forecasting models to the time series and determining a respective champion forecast for the time series by comparing error metrics associated with the respective forecasts. Each of the model strategies is further configured to determine error values associated with the champion forecasts, each of the error values indicating the accuracy of a respective champion forecast among the champion forecasts, and generate one of the error distributions using the error values. The sequence of operations also includes a model-strategy comparison operation, which is configured to receive the error distributions corresponding to the model strategies, determine aggregate error values associated with the model strategies based on the error distributions, each aggregate error value being associated with one of the model strategies and being determined by aggregating the error values in one of the error distributions, and determine which of the model strategies is a champion model strategy for the plurality of time series by comparing the aggregate error values to one another. The instructions can cause the processing device to execute the pipeline to determine the champion model strategy for the plurality of time series.
This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used in isolation to determine the scope of the claimed subject matter. The subject matter should be understood by reference to appropriate portions of the entire specification, any or all drawings, and each claim.
The foregoing, together with other features and examples, will become more apparent upon referring to the following specification, claims, and accompanying drawings.
The present disclosure is described in conjunction with the appended figures:
In the appended figures, similar components or features can have the same reference label. Further, various components of the same type can be distinguished by following the reference label by a dash and a second label that distinguishes among the similar components. If only the first reference label is used in the specification, the description is applicable to any one of the similar components having the same first reference label irrespective of the second reference label.
Time series forecasting involves using a model to predict future values for a time series. Existing time series forecasting systems are rigid and closed. The forecasting process uses one set of modeling properties with little opportunity to specify alternative modeling approaches. Some examples of the present disclosure involve a pipeline system for time series forecasting that can overcome these problems by providing mechanisms for building and executing one or more pipelines for time series forecasting. These mechanisms can include a flexible, user-friendly, graphical user interface (GUI) components that allows complicated forecasting pipelines and projects to be built and executed efficiently, while circumventing the arduous process of programming, testing, and implementing complex commands.
Specifically, some examples of the present disclosure include a graphical user interface (GUI) in which a pipeline can be built by dragging and dropping user interface components representing different operations in a pipeline or by importing an existing pipeline. For example, the GUI can provide a library of model strategies from which a user can select and position an appropriate model strategy in the pipeline. The pipeline system can also automatically add operations to, or remove operations from, the pipeline depending on the user's selections. Examples of such operations can include pre-processing, model strategy comparison, and pipeline segmentation operations. Automatically adding and removing dependent operations can prevent failures and inaccuracies. The pipeline system can store the pipelines in files, which can be subsequently edited, copied, and transferred among users.
As described herein, a “pipeline” or a “forecast pipeline” includes a preset, repeatable sequence of operations (e.g., an end-to-end sequence of operations) for processing time series data to produce forecasts for the time series data. The sequence of operations forming a pipeline are defined and stored in a transferrable file so that various users can share and implement the pipeline. In some examples, pipelines can be generated via a graphical user interface (GUI) through which users can easily create, delete, edit, import, share, and execute them. Pipelines may be executed by a group of services (e.g., microservices) in a distributed computing environment, where each service can implement at least one aspect of the pipeline's functionality.
As one specific example, a pipeline can include an end-to-end sequence of operations for forecasting a set of time series. The operations can include a data preprocessing operation to prepare the set of time series for forecasting. The operations can also include a model strategy operations for applying various model strategies to the set of time series. The operations can further include strategy comparison operations for comparing the various model strategies to determine a champion model strategy. And the operations can include an output operation for generating a champion forecast for the set of time series based on the champion model strategy. Each of these operations is further described in turn below.
A data preprocessing operation processes the set of time series before the forecasting is perform so that the format, content or other aspects of the set of time series are compatible with the forecasting process. The data preprocessing operation can include, for example, normalizing, cleaning, adding, removing, or reformatting the set of time-series data.
A model strategy is a process defining how the set of time series is to be modeled, such as by applying different forecasting models (e.g., ARIMA, ARIMAX, or ESM) to the set of time series. The model strategy can be defined in shareable files to enable importation of the model strategy into another pipeline for execution. In some examples, a model strategy can involve determining champion forecasts for the set of time series. For each time series in the set, forecasts for the time series are generated by applying forecasting models described in the model strategy to the time series and a champion forecast for the time series can be determined by comparing error metrics associated with these forecasts. An error distribution can be generated for the model strategy using the error metrics associated with a corresponding champion forecast.
A model-strategy comparison operation compares the model strategies based on the error distributions for the model strategies to determine which of the model strategies is a champion model strategy for the time series. As used herein, a “champion model strategy” for a set of time series refers to a model strategy that produces the most accurate forecasts overall for the set of time series, measured by, for example, the smallest aggregate error value among the multiple model strategies.
An output operation produces an output containing at least some of the data generated by the pipeline, such as the champion forecasts generated by applying the champion model strategy to the set of time series.
Some examples of the present disclosure improve the time series forecasting by providing a tool that allows complicated pipelines and projects to be built and executed efficiently using flexible, user-friendly user interface components that avoid the arduous process of programming, testing, and implementing complex commands. This tool allows multiple alternative modeling strategies to be generated for multiple time series and compared simultaneously (or near simultaneously) to select a champion model strategy. This significantly improves the efficiency and flexibility of the time series forecasting system. In addition, the model strategy generated in one pipeline can be saved and reused in another pipeline for the same set of time series or a different set of time series. Likewise, the pipeline generated for a set of time series can be saved and reused for another set of time series. The reusability of the model strategy and the pipeline significantly increases the flexibility of the forecasting system and also the efficiency of performing a forecasting task.
These illustrative examples are given to introduce the reader to the general subject matter discussed here and are not intended to limit the scope of the disclosed concepts. The following sections describe various additional features and examples with reference to the drawings in which like numerals indicate like elements but, like the illustrative examples, should not be used to limit the present disclosure.
Data transmission network 100 may also include computing environment 114. Computing environment 114 may be a specialized computer or other machine that processes the data received within the data transmission network 100. The computing environment 114 may include one or more other systems. For example, computing environment 114 may include a database system 118 or a communications grid 120. The computing environment 114 can include one or more processing devices (e.g., distributed over one or more networks or otherwise in communication with one another) that, in some examples, can collectively be referred to as a processor or a processing device.
Data transmission network 100 also includes one or more network devices 102. Network devices 102 may include client devices that can communicate with computing environment 114. For example, network devices 102 may send data to the computing environment 114 to be processed, may send communications to the computing environment 114 to control different aspects of the computing environment or the data it is processing, among other reasons. Network devices 102 may interact with the computing environment 114 through a number of ways, such as, for example, over one or more networks 108.
In some examples, network devices 102 may provide a large amount of data, either all at once or streaming over a period of time (e.g., using event stream processing (ESP)), to the computing environment 114 via networks 108. For example, the network devices 102 can transmit electronic messages for use in executing one or more forecast pipelines, all at once or streaming over a period of time, to the computing environment 114 via networks 108.
The network devices 102 may include network computers, sensors, databases, or other devices that may transmit or otherwise provide data to computing environment 114. For example, network devices 102 may include local area network devices, such as routers, hubs, switches, or other computer networking devices. These devices may provide a variety of stored or generated data, such as network data or data specific to the network devices 102 themselves. Network devices 102 may also include sensors that monitor their environment or other devices to collect data regarding that environment or those devices, and such network devices 102 may provide data they collect over time. Network devices 102 may also include devices within the internet of things, such as devices within a home automation network. Some of these devices may be referred to as edge devices, and may involve edge-computing circuitry. Data may be transmitted by network devices 102 directly to computing environment 114 or to network-attached data stores, such as network-attached data stores 110 for storage so that the data may be retrieved later by the computing environment 114 or other portions of data transmission network 100. For example, the network devices 102 can transmit data usable for executing one or more forecast pipelines to a network-attached data store 110 for storage. The computing environment 114 may later retrieve the data from the network-attached data store 110 and use the data to facilitate executing forecast pipelines.
Network-attached data stores 110 can store data to be processed by the computing environment 114 as well as any intermediate or final data generated by the computing system in non-volatile memory. But in certain examples, the configuration of the computing environment 114 allows its operations to be performed such that intermediate and final data results can be stored solely in volatile memory (e.g., RAM), without a requirement that intermediate or final data results be stored to non-volatile types of memory (e.g., disk). This can be useful in certain situations, such as when the computing environment 114 receives ad hoc queries from a user and when responses, which are generated by processing large amounts of data, need to be generated dynamically (e.g., on the fly). In this situation, the computing environment 114 may be configured to retain the processed information within memory so that responses can be generated for the user at different levels of detail as well as allow a user to interactively query against this information.
Network-attached data stores 110 may store a variety of different types of data organized in a variety of different ways and from a variety of different sources. For example, network-attached data stores may include storage other than primary storage located within computing environment 114 that is directly accessible by processors located therein. Network-attached data stores may include secondary, tertiary or auxiliary storage, such as large hard drives, servers, virtual memory, among other types. Storage devices may include portable or non-portable storage devices, optical storage devices, and various other mediums capable of storing, containing data. A machine-readable storage medium or computer-readable storage medium may include a non-transitory medium in which data can be stored and that does not include carrier waves or transitory electronic communications. Examples of a non-transitory medium may include, for example, a magnetic disk or tape, optical storage media such as compact disk or digital versatile disk, flash memory, memory or memory devices. A computer-program product may include code or machine-executable instructions that may represent a procedure, a function, a subprogram, a program, a routine, a subroutine, a module, a software package, a class, or any combination of instructions, data structures, or program statements. A code segment may be coupled to another code segment or a hardware circuit by passing or receiving information, data, arguments, parameters, or memory contents. Information, arguments, parameters, data, etc. may be passed, forwarded, or transmitted via any suitable means including memory sharing, message passing, token passing, network transmission, among others. Furthermore, the data stores may hold a variety of different types of data. For example, network-attached data stores 110 may hold unstructured (e.g., raw) data.
The unstructured data may be presented to the computing environment 114 in different forms such as a flat file or a conglomerate of data records, and may have data values and accompanying time stamps. The computing environment 114 may be used to analyze the unstructured data in a variety of ways to determine the best way to structure (e.g., hierarchically) that data, such that the structured data is tailored to a type of further analysis that a user wishes to perform on the data. For example, after being processed, the unstructured time-stamped data may be aggregated by time (e.g., into daily time period units) to generate time-series data or structured hierarchically according to one or more dimensions (e.g., parameters, attributes, or variables). For example, data may be stored in a hierarchical data structure, such as a relational online analytical processing (ROLAP) or multidimensional online analytical processing (MOLAP) database, or may be stored in another tabular form, such as in a flat-hierarchy form.
Data transmission network 100 may also include one or more server farms 106. Computing environment 114 may route select communications or data to the sever farms 106 or one or more servers within the server farms 106. Server farms 106 can be configured to provide information in a predetermined manner. For example, server farms 106 may access data to transmit in response to a communication. Server farms 106 may be separately housed from each other device within data transmission network 100, such as computing environment 114, or may be part of a device or system.
Server farms 106 may host a variety of different types of data processing as part of data transmission network 100. Server farms 106 may receive a variety of different data from network devices, from computing environment 114, from cloud network 116, or from other sources. The data may have been obtained or collected from one or more websites, sensors, as inputs from a control database, or may have been received as inputs from an external system or device. Server farms 106 may assist in processing the data by turning raw data into processed data based on one or more rules implemented by the server farms. For example, sensor data may be analyzed to determine changes in an environment over time or in real-time.
Data transmission network 100 may also include one or more cloud networks 116. Cloud network 116 may include a cloud infrastructure system that provides cloud services. In certain examples, services provided by the cloud network 116 may include a host of services that are made available to users of the cloud infrastructure system on demand. Cloud network 116 is shown in
While each device, server, and system in
Each communication within data transmission network 100 (e.g., between client devices, between a device and connection management system 150, between server farms 106 and computing environment 114, or between a server and a device) may occur over one or more networks 108. Networks 108 may include one or more of a variety of different types of networks, including a wireless network, a wired network, or a combination of a wired and wireless network. Examples of suitable networks include the Internet, a personal area network, a local area network (LAN), a wide area network (WAN), or a wireless local area network (WLAN). A wireless network may include a wireless interface or combination of wireless interfaces. As an example, a network in the one or more networks 108 may include a short-range communication channel, such as a Bluetooth or a Bluetooth Low Energy channel. A wired network may include a wired interface. The wired or wireless networks may be implemented using routers, access points, bridges, gateways, or the like, to connect devices in the network 108. The networks 108 can be incorporated entirely within or can include an intranet, an extranet, or a combination thereof. In one example, communications between two or more systems or devices can be achieved by a secure communications protocol, such as secure sockets layer (SSL) or transport layer security (TLS). In addition, data or transactional details may be encrypted.
Some aspects may utilize the Internet of Things (IoT), where things (e.g., machines, devices, phones, sensors) can be connected to networks and the data from these things can be collected and processed within the things or external to the things. For example, the IoT can include sensors in many different devices, and high value analytics can be applied to identify hidden relationships and drive increased efficiencies. This can apply to both big data analytics and real-time (e.g., ESP) analytics.
As noted, computing environment 114 may include a communications grid 120 and a transmission network database system 118. Communications grid 120 may be a grid-based computing system for processing large amounts of data. The transmission network database system 118 may be for managing, storing, and retrieving large amounts of data that are distributed to and stored in the one or more network-attached data stores 110 or other data stores that reside at different locations within the transmission network database system 118. The computing nodes in the communications grid 120 and the transmission network database system 118 may share the same processor hardware, such as processors that are located within computing environment 114.
In some examples, the computing environment 114, a network device 102, or both can implement one or more processes for executing forecast pipelines to generate forecasts for time-series data. For example, the computing environment 114, a network device 102, or both can implement one or more versions of the processes discussed with respect to any of the figures.
As shown in
Although network devices 204-209 are shown in
The network devices 204-209 may also perform processing on data it collects before transmitting the data to the computing environment 214, or before deciding whether to transmit data to the computing environment 214. For example, network devices 204-209 may determine whether data collected meets certain rules, for example by comparing data or values calculated from the data and comparing that data to one or more thresholds. The network devices 204-209 may use this data or comparisons to determine if the data is to be transmitted to the computing environment 214 for further use or processing. In some examples, the network devices 204-209 can pre-process the data prior to transmitting the data to the computing environment 214. For example, the network devices 204-209 can reformat the data before transmitting the data to the computing environment 214 for further processing (e.g., analyzing the data to facilitate the execution of forecast pipelines).
Computing environment 214 may include machines 220, 240. Although computing environment 214 is shown in
Computing environment 214 can communicate with various devices via one or more routers 225 or other inter-network or intra-network connection components. For example, computing environment 214 may communicate with client devices 230 via one or more routers 225. Computing environment 214 may collect, analyze or store data from or pertaining to communications, client device operations, client rules, or user-associated actions stored at one or more data stores 235. Such data may influence communication routing to the devices within computing environment 214, how data is stored or processed within computing environment 214, among other actions.
Notably, various other devices can further be used to influence communication routing or processing between devices within computing environment 214 and with devices outside of computing environment 214. For example, as shown in
In addition to computing environment 214 collecting data (e.g., as received from network devices, such as sensors, and client devices or other sources) to be processed as part of a big data analytics project, it may also receive data in real time as part of a streaming analytics environment. As noted, data may be collected using a variety of sources as communicated via different kinds of networks or locally. Such data may be received on a real-time streaming basis. For example, network devices 204-209 may receive data periodically and in real time from a web server or other source. Devices within computing environment 214 may also perform pre-analysis on data it receives to determine if the data received should be processed as part of an ongoing project. For example, as part of a project in which forecasts are to be generated for time-series data, the computing environment 214 can perform a pre-analysis of the data. The pre-analysis can include determining whether the data is in a correct format for executing the forecast pipelines using the data and, if not, reformatting the data into the correct format.
The model 300 can include layers 302-314. The layers 302-314 are arranged in a stack. Each layer in the stack serves the layer one level higher than it (except for the application layer, which is the highest layer), and is served by the layer one level below it (except for the physical layer 302, which is the lowest layer). The physical layer 302 is the lowest layer because it receives and transmits raw bites of data, and is the farthest layer from the user in a communications system. On the other hand, the application layer is the highest layer because it interacts directly with a software application.
As noted, the model 300 includes a physical layer 302. Physical layer 302 represents physical communication, and can define parameters of that physical communication. For example, such physical communication may come in the form of electrical, optical, or electromagnetic communications. Physical layer 302 also defines protocols that may control communications within a data transmission network.
Link layer 304 defines links and mechanisms used to transmit (e.g., move) data across a network. The link layer manages node-to-node communications, such as within a grid-computing environment. Link layer 304 can detect and correct errors (e.g., transmission errors in the physical layer 302). Link layer 304 can also include a media access control (MAC) layer and logical link control (LLC) layer.
Network layer 306 can define the protocol for routing within a network. In other words, the network layer coordinates transferring data across nodes in a same network (e.g., such as a grid-computing environment). Network layer 306 can also define the processes used to structure local addressing within the network.
Transport layer 308 can manage the transmission of data and the quality of the transmission or receipt of that data. Transport layer 308 can provide a protocol for transferring data, such as, for example, a Transmission Control Protocol (TCP). Transport layer 308 can assemble and disassemble data frames for transmission. The transport layer can also detect transmission errors occurring in the layers below it.
Session layer 310 can establish, maintain, and manage communication connections between devices on a network. In other words, the session layer controls the dialogues or nature of communications between network devices on the network. The session layer may also establish checkpointing, adjournment, termination, and restart procedures.
Presentation layer 312 can provide translation for communications between the application and network layers. In other words, this layer may encrypt, decrypt or format data based on data types known to be accepted by an application or network layer.
Application layer 314 interacts directly with software applications and end users, and manages communications between them. Application layer 314 can identify destinations, local resource states or availability or communication content or formatting using the applications.
For example, a communication link can be established between two devices on a network. One device can transmit an analog or digital representation of an electronic message that includes a data set to the other device. The other device can receive the analog or digital representation at the physical layer 302. The other device can transmit the data associated with the electronic message through the remaining layers 304-314. The application layer 314 can receive data associated with the electronic message. The application layer 314 can identify one or more applications, such as an application for executing forecast pipelines, to which to transmit data associated with the electronic message. The application layer 314 can transmit the data to the identified application.
Intra-network connection components 322, 324 can operate in lower levels, such as physical layer 302 and link layer 304, respectively. For example, a hub can operate in the physical layer, a switch can operate in the physical layer, and a router can operate in the network layer. Inter-network connection components 326, 328 are shown to operate on higher levels, such as layers 306-314. For example, routers can operate in the network layer and network devices can operate in the transport, session, presentation, and application layers.
A computing environment 330 can interact with or operate on, in various examples, one, more, all or any of the various layers. For example, computing environment 330 can interact with a hub (e.g., via the link layer) to adjust which devices the hub communicates with. The physical layer 302 may be served by the link layer 304, so it may implement such data from the link layer 304. For example, the computing environment 330 may control which devices from which it can receive data. For example, if the computing environment 330 knows that a certain network device has turned off, broken, or otherwise become unavailable or unreliable, the computing environment 330 may instruct the hub to prevent any data from being transmitted to the computing environment 330 from that network device. Such a process may be beneficial to avoid receiving data that is inaccurate or that has been influenced by an uncontrolled environment. As another example, computing environment 330 can communicate with a bridge, switch, router or gateway and influence which device within the system (e.g., system 200) the component selects as a destination. In some examples, computing environment 330 can interact with various layers by exchanging communications with equipment operating on a particular layer by routing or modifying existing communications. In another example, such as in a grid-computing environment, a node may determine how data within the environment should be routed (e.g., which node should receive certain data) based on certain parameters or information provided by other layers within the model.
The computing environment 330 may be a part of a communications grid environment, the communications of which may be implemented as shown in the protocol of
Communications grid computing system 400 (which can be referred to as a “communications grid”) also includes one or more worker nodes. Shown in
A control node 402-406 may connect with an external device with which the control node 402-406 may communicate (e.g., a communications grid user, such as a server or computer, may connect to a controller of the grid). For example, a server or computer may connect to control nodes 402-406 and may transmit a project or job to the node, such as a project or job related to executing forecast pipelines to generate forecasts for time-series data. The project may include the data set. The data set may be of any size. Once the control node 402-406 receives such a project including a large data set, the control node may distribute the data set or projects related to the data set to be performed by worker nodes. Alternatively, for a project including a large data set, the data set may be receive or stored by a machine other than a control node 402-406 (e.g., a Hadoop data node).
Control nodes 402-406 can maintain knowledge of the status of the nodes in the grid (e.g., grid status information), accept work requests from clients, subdivide the work across worker nodes, and coordinate the worker nodes, among other responsibilities. Worker nodes 412-420 may accept work requests from a control node 402-406 and provide the control node with results of the work performed by the worker node. A grid may be started from a single node (e.g., a machine, computer, server, etc.). This first node may be assigned or may start as the primary control node 402 that will control any additional nodes that enter the grid.
When a project is submitted for execution (e.g., by a client or a controller of the grid) it may be assigned to a set of nodes. After the nodes are assigned to a project, a data structure (e.g., a communicator) may be created. The communicator may be used by the project for information to be shared between the project code running on each node. A communication handle may be created on each node. A handle, for example, is a reference to the communicator that is valid within a single process on a single node, and the handle may be used when requesting communications between nodes.
A control node, such as control node 402, may be designated as the primary control node. A server, computer or other external device may connect to the primary control node. Once the control node 402 receives a project, the primary control node may distribute portions of the project to its worker nodes for execution. For example, a project for executing forecast pipelines to generate forecasts for time-series data can be initiated on communications grid computing system 400. A primary control node can control the work to be performed for the project in order to complete the project as requested or instructed. The primary control node may distribute work to the worker nodes 412-420 based on various factors, such as which subsets or portions of projects may be completed most efficiently and in the correct amount of time. For example, a worker node 412 may execute executing software modules involved in the forecast pipelines using at least a portion of data that is already local (e.g., stored on) the worker node. The primary control node also coordinates and processes the results of the work performed by each worker node 412-420 after each worker node 412-420 executes and completes its job. For example, the primary control node may receive a result from one or more worker nodes 412-420, and the primary control node may organize (e.g., collect and assemble) the results received and compile them to produce a complete result for the project received from the end user.
Any remaining control nodes, such as control nodes 404, 406, may be assigned as backup control nodes for the project. In an example, backup control nodes may not control any portion of the project. Instead, backup control nodes may serve as a backup for the primary control node and take over as primary control node if the primary control node were to fail. If a communications grid were to include only a single control node 402, and the control node 402 were to fail (e.g., the control node is shut off or breaks) then the communications grid as a whole may fail and any project or job being run on the communications grid may fail and may not complete. While the project may be run again, such a failure may cause a delay (severe delay in some cases, such as overnight delay) in completion of the project. Therefore, a grid with multiple control nodes 402-406, including a backup control node, may be beneficial.
In some examples, the primary control node may open a pair of listening sockets to add another node or machine to the grid. A socket may be used to accept work requests from clients, and the second socket may be used to accept connections from other grid nodes. The primary control node may be provided with a list of other nodes (e.g., other machines, computers, servers, etc.) that can participate in the grid, and the role that each node can fill in the grid. Upon startup of the primary control node (e.g., the first node on the grid), the primary control node may use a network protocol to start the server process on every other node in the grid. Command line parameters, for example, may inform each node of one or more pieces of information, such as: the role that the node will have in the grid, the host name of the primary control node, the port number on which the primary control node is accepting connections from peer nodes, among others. The information may also be provided in a configuration file, transmitted over a secure shell tunnel, recovered from a configuration server, among others. While the other machines in the grid may not initially know about the configuration of the grid, that information may also be sent to each other node by the primary control node. Updates of the grid information may also be subsequently sent to those nodes.
For any control node other than the primary control node added to the grid, the control node may open three sockets. The first socket may accept work requests from clients, the second socket may accept connections from other grid members, and the third socket may connect (e.g., permanently) to the primary control node. When a control node (e.g., primary control node) receives a connection from another control node, it first checks to see if the peer node is in the list of configured nodes in the grid. If it is not on the list, the control node may clear the connection. If it is on the list, it may then attempt to authenticate the connection. If authentication is successful, the authenticating node may transmit information to its peer, such as the port number on which a node is listening for connections, the host name of the node, information about how to authenticate the node, among other information. When a node, such as the new control node, receives information about another active node, it can check to see if it already has a connection to that other node. If it does not have a connection to that node, it may then establish a connection to that control node.
Any worker node added to the grid may establish a connection to the primary control node and any other control nodes on the grid. After establishing the connection, it may authenticate itself to the grid (e.g., any control nodes, including both primary and backup, or a server or user controlling the grid). After successful authentication, the worker node may accept configuration information from the control node.
When a node joins a communications grid (e.g., when the node is powered on or connected to an existing node on the grid or both), the node is assigned (e.g., by an operating system of the grid) a universally unique identifier (UUID). This unique identifier may help other nodes and external entities (devices, users, etc.) to identify the node and distinguish it from other nodes. When a node is connected to the grid, the node may share its unique identifier with the other nodes in the grid. Since each node may share its unique identifier, each node may know the unique identifier of every other node on the grid. Unique identifiers may also designate a hierarchy of each of the nodes (e.g., backup control nodes) within the grid. For example, the unique identifiers of each of the backup control nodes may be stored in a list of backup control nodes to indicate an order in which the backup control nodes will take over for a failed primary control node to become a new primary control node. But, a hierarchy of nodes may also be determined using methods other than using the unique identifiers of the nodes. For example, the hierarchy may be predetermined, or may be assigned based on other predetermined factors.
The grid may add new machines at any time (e.g., initiated from any control node). Upon adding a new node to the grid, the control node may first add the new node to its table of grid nodes. The control node may also then notify every other control node about the new node. The nodes receiving the notification may acknowledge that they have updated their configuration information.
Primary control node 402 may, for example, transmit one or more communications to backup control nodes 404, 406 (and, for example, to other control or worker nodes 412-420 within the communications grid). Such communications may be sent periodically, at fixed time intervals, between known fixed stages of the project's execution, among other protocols. The communications transmitted by primary control node 402 may be of varied types and may include a variety of types of information. For example, primary control node 402 may transmit snapshots (e.g., status information) of the communications grid so that backup control node 404 always has a recent snapshot of the communications grid. The snapshot or grid status may include, for example, the structure of the grid (including, for example, the worker nodes 410-420 in the communications grid, unique identifiers of the worker nodes 410-420, or their relationships with the primary control node 402) and the status of a project (including, for example, the status of each worker node's portion of the project). The snapshot may also include analysis or results received from worker nodes 410-420 in the communications grid. The backup control nodes 404, 406 may receive and store the backup data received from the primary control node 402. The backup control nodes 404, 406 may transmit a request for such a snapshot (or other information) from the primary control node 402, or the primary control node 402 may send such information periodically to the backup control nodes 404, 406.
As noted, the backup data may allow a backup control node 404, 406 to take over as primary control node if the primary control node 402 fails without requiring the communications grid to start the project over from scratch. If the primary control node 402 fails, the backup control node 404, 406 that will take over as primary control node may retrieve the most recent version of the snapshot received from the primary control node 402 and use the snapshot to continue the project from the stage of the project indicated by the backup data. This may prevent failure of the project as a whole.
A backup control node 404, 406 may use various methods to determine that the primary control node 402 has failed. In one example of such a method, the primary control node 402 may transmit (e.g., periodically) a communication to the backup control node 404, 406 that indicates that the primary control node 402 is working and has not failed, such as a heartbeat communication. The backup control node 404, 406 may determine that the primary control node 402 has failed if the backup control node has not received a heartbeat communication for a certain predetermined period of time. Alternatively, a backup control node 404, 406 may also receive a communication from the primary control node 402 itself (before it failed) or from a worker node 410-420 that the primary control node 402 has failed, for example because the primary control node 402 has failed to communicate with the worker node 410-420.
Different methods may be performed to determine which backup control node of a set of backup control nodes (e.g., backup control nodes 404, 406) can take over for failed primary control node 402 and become the new primary control node. For example, the new primary control node may be chosen based on a ranking or “hierarchy” of backup control nodes based on their unique identifiers. In an alternative example, a backup control node may be assigned to be the new primary control node by another device in the communications grid or from an external device (e.g., a system infrastructure or an end user, such as a server or computer, controlling the communications grid). In another alternative example, the backup control node that takes over as the new primary control node may be designated based on bandwidth or other statistics about the communications grid.
A worker node within the communications grid may also fail. If a worker node fails, work being performed by the failed worker node may be redistributed amongst the operational worker nodes. In an alternative example, the primary control node may transmit a communication to each of the operable worker nodes still on the communications grid that each of the worker nodes should purposefully fail also. After each of the worker nodes fail, they may each retrieve their most recent saved checkpoint of their status and re-start the project from that checkpoint to minimize lost progress on the project being executed. In some examples, a communications grid computing system 400 can be used to execute the operations involved in the forecast pipelines.
The process may also include receiving a failure communication corresponding to a node in the communications grid in operation 506. For example, a node may receive a failure communication including an indication that the primary control node has failed, prompting a backup control node to take over for the primary control node. In an alternative embodiment, a node may receive a failure that a worker node has failed, prompting a control node to reassign the work being performed by the worker node. The process may also include reassigning a node or a portion of the project being executed by the failed node, as described in operation 508. For example, a control node may designate the backup control node as a new primary control node based on the failure communication upon receiving the failure communication. If the failed node is a worker node, a control node may identify a project status of the failed worker node using the snapshot of the communications grid, where the project status of the failed worker node includes a status of a portion of the project being executed by the failed worker node at the failure time.
The process may also include receiving updated grid status information based on the reassignment, as described in operation 510, and transmitting a set of instructions based on the updated grid status information to one or more nodes in the communications grid, as described in operation 512. The updated grid status information may include an updated project status of the primary control node or an updated project status of the worker node. The updated information may be transmitted to the other nodes in the grid to update their stale stored information.
Similar to in
Each node also includes a data store 624. Data stores 624, similar to network-attached data stores 110 in
Each node also includes a user-defined function (UDF) 626. The UDF provides a mechanism for the DMBS 628 to transfer data to or receive data from the database stored in the data stores 624 that are managed by the DBMS. For example, UDF 626 can be invoked by the DBMS to provide data to the GESC for processing. The UDF 626 may establish a socket connection (not shown) with the GESC to transfer the data. Alternatively, the UDF 626 can transfer data to the GESC by writing data to shared memory accessible by both the UDF and the GESC.
The GESC 620 at the nodes 602 and 610 may be connected via a network, such as network 108 shown in
DMBS 628 may control the creation, maintenance, and use of database or data structure (not shown) within nodes 602 or 610. The database may organize data stored in data stores 624. The DMBS 628 at control node 602 may accept requests for data and transfer the appropriate data for the request. With such a process, collections of data may be distributed across multiple physical locations. In this example, each node 602 and 610 stores a portion of the total data managed by the management system in its associated data store 624.
Furthermore, the DBMS may be responsible for protecting against data loss using replication techniques. Replication includes providing a backup copy of data stored on one node on one or more other nodes. Therefore, if one node fails, the data from the failed node can be recovered from a replicated copy residing at another node. However, as described herein with respect to
To initiate the project, the control node may determine if the query requests use of the grid-based computing environment to execute the project. If the determination is no, then the control node initiates execution of the project in a solo environment (e.g., at the control node), as described in operation 710. If the determination is yes, the control node may initiate execution of the project in the grid-based computing environment, as described in operation 706. In such a situation, the request may include a requested configuration of the grid. For example, the request may include a number of control nodes and a number of worker nodes to be used in the grid when executing the project. After the project has been completed, the control node may transmit results of the analysis yielded by the grid, as described in operation 708. Whether the project is executed in a solo or grid-based environment, the control node provides the results of the project.
As noted with respect to
The ESPE may receive streaming data over a period of time related to certain events, such as events or other data sensed by one or more network devices. The ESPE may perform operations associated with processing data created by the one or more devices. For example, the ESPE may receive data from the one or more network devices 204-209 shown in
The engine container is the top-level container in a model that manages the resources of the one or more projects 802. In an illustrative example, there may be only one ESPE 800 for each instance of the ESP application, and ESPE 800 may have a unique engine name. Additionally, the one or more projects 802 may each have unique project names, and each query may have a unique continuous query name and begin with a uniquely named source window of the one or more source windows 806. ESPE 800 may or may not be persistent.
Continuous query modeling involves defining directed graphs of windows for event stream manipulation and transformation. A window in the context of event stream manipulation and transformation is a processing node in an event stream processing model. A window in a continuous query can perform aggregations, computations, pattern-matching, and other operations on data flowing through the window. A continuous query may be described as a directed graph of source, relational, pattern matching, and procedural windows. The one or more source windows 806 and the one or more derived windows 808 represent continuously executing queries that generate updates to a query result set as new event blocks stream through ESPE 800. A directed graph, for example, is a set of nodes connected by edges, where the edges have a direction associated with them.
An event object may be described as a packet of data accessible as a collection of fields, with at least one of the fields defined as a key or unique identifier (ID). The event object may be created using a variety of formats including binary, alphanumeric, XML, etc. Each event object may include one or more fields designated as a primary identifier (ID) for the event so ESPE 800 can support operation codes (opcodes) for events including insert, update, upsert, and delete. Upsert opcodes update the event if the key field already exists; otherwise, the event is inserted. For illustration, an event object may be a packed binary representation of a set of field values and include both metadata and field data associated with an event. The metadata may include an opcode indicating if the event represents an insert, update, delete, or upsert, a set of flags indicating if the event is a normal, partial-update, or a retention generated event from retention policy management, and a set of microsecond timestamps that can be used for latency measurements.
An event block object may be described as a grouping or package of event objects. An event stream may be described as a flow of event block objects. A continuous query of the one or more continuous queries 804 transforms a source event stream made up of streaming event block objects published into ESPE 800 into one or more output event streams using the one or more source windows 806 and the one or more derived windows 808. A continuous query can also be thought of as data flow modeling.
The one or more source windows 806 are at the top of the directed graph and have no windows feeding into them. Event streams are published into the one or more source windows 806, and from there, the event streams may be directed to the next set of connected windows as defined by the directed graph. The one or more derived windows 808 are all instantiated windows that are not source windows and that have other windows streaming events into them. The one or more derived windows 808 may perform computations or transformations on the incoming event streams. The one or more derived windows 808 transform event streams based on the window type (that is operators such as join, filter, compute, aggregate, copy, pattern match, procedural, union, etc.) and window settings. As event streams are published into ESPE 800, they are continuously queried, and the resulting sets of derived windows in these queries are continuously updated.
Within the application, a user may interact with one or more user interface windows presented to the user in a display under control of the ESPE independently or through a browser application in an order selectable by the user. For example, a user may execute an ESP application, which causes presentation of a first user interface window, which may include a plurality of menus and selectors such as drop down menus, buttons, text boxes, hyperlinks, etc. associated with the ESP application as understood by a person of skill in the art. Various operations may be performed in parallel, for example, using a plurality of threads.
At operation 900, an ESP application may define and start an ESPE, thereby instantiating an ESPE at a device, such as machine 220 and/or 240. In an operation 902, the engine container is created. For illustration, ESPE 800 may be instantiated using a function call that specifies the engine container as a manager for the model.
In an operation 904, the one or more continuous queries 804 are instantiated by ESPE 800 as a model. The one or more continuous queries 804 may be instantiated with a dedicated thread pool or pools that generate updates as new events stream through ESPE 800. For illustration, the one or more continuous queries 804 may be created to model business processing logic within ESPE 800, to predict events within ESPE 800, to model a physical system within ESPE 800, to predict the physical system state within ESPE 800, etc. For example, as noted, ESPE 800 may be used to support sensor data monitoring and management (e.g., sensing may include force, torque, load, strain, position, temperature, air pressure, fluid flow, chemical properties, resistance, electromagnetic fields, radiation, irradiance, proximity, acoustics, moisture, distance, speed, vibrations, acceleration, electrical potential, or electrical current, etc.).
ESPE 800 may analyze and process events in motion or “event streams.” Instead of storing data and running queries against the stored data, ESPE 800 may store queries and stream data through them to allow continuous analysis of data as it is received. The one or more source windows 806 and the one or more derived windows 808 may be created based on the relational, pattern matching, and procedural algorithms that transform the input event streams into the output event streams to model, simulate, score, test, predict, etc. based on the continuous query model defined and application to the streamed data.
In an operation 906, a publish/subscribe (pub/sub) capability is initialized for ESPE 800. In an illustrative embodiment, a pub/sub capability is initialized for each project of the one or more projects 802. To initialize and enable pub/sub capability for ESPE 800, a port number may be provided. Pub/sub clients can use a host name of an ESP device running the ESPE and the port number to establish pub/sub connections to ESPE 800.
Publish-subscribe is a message-oriented interaction paradigm based on indirect addressing. Processed data recipients specify their interest in receiving information from ESPE 800 by subscribing to specific classes of events, while information sources publish events to ESPE 800 without directly addressing the receiving parties. ESPE 800 coordinates the interactions and processes the data. In some cases, the data source receives confirmation that the published information has been received by a data recipient.
A publish/subscribe API may be described as a library that enables an event publisher, such as publishing device 1022, to publish event streams into ESPE 800 or an event subscriber, such as event subscription device A 1024a, event subscription device B 1024b, and event subscription device C 1024c, to subscribe to event streams from ESPE 800. For illustration, one or more publish/subscribe APIs may be defined. Using the publish/subscribe API, an event publishing application may publish event streams into a running event stream processor project source window of ESPE 800, and the event subscription application may subscribe to an event stream processor project source window of ESPE 800.
The publish/subscribe API provides cross-platform connectivity and endianness compatibility between ESP application and other networked applications, such as event publishing applications instantiated at publishing device 1022, and event subscription applications instantiated at one or more of event subscription device A 1024a, event subscription device B 1024b, and event subscription device C 1024c.
Referring back to
ESP subsystem 1001 may include a publishing client 1002, ESPE 800, a subscribing client A 1004, a subscribing client B 1006, and a subscribing client C 1008. Publishing client 1002 may be started by an event publishing application executing at publishing device 1022 using the publish/subscribe API. Subscribing client A 1004 may be started by an event subscription application A, executing at event subscription device A 1024a using the publish/subscribe API. Subscribing client B 1006 may be started by an event subscription application B executing at event subscription device B 1024b using the publish/subscribe API. Subscribing client C 1008 may be started by an event subscription application C executing at event subscription device C 1024c using the publish/subscribe API.
An event block object containing one or more event objects is injected into a source window of the one or more source windows 806 from an instance of an event publishing application on publishing device 1022. The event block object may be generated, for example, by the event publishing application and may be received by publishing client 1002. A unique ID may be maintained as the event block object is passed between the one or more source windows 806 and/or the one or more derived windows 808 of ESPE 800, and to subscribing client A 1004, subscribing client B 1006, and subscribing client C 1008 and to event subscription device A 1024a, event subscription device B 1024b, and event subscription device C 1024c. Publishing client 1002 may further generate and include a unique embedded transaction ID in the event block object as the event block object is processed by a continuous query, as well as the unique ID that publishing device 1022 assigned to the event block object.
In an operation 912, the event block object is processed through the one or more continuous queries 804. In an operation 914, the processed event block object is output to one or more computing devices of the event subscription devices 1024a-c. For example, subscribing client A 1004, subscribing client B 1006, and subscribing client C 1008 may send the received event block object to event subscription device A 1024a, event subscription device B 1024b, and event subscription device C 1024c, respectively.
ESPE 800 maintains the event block containership aspect of the received event blocks from when the event block is published into a source window and works its way through the directed graph defined by the one or more continuous queries 804 with the various event translations before being output to subscribers. Subscribers can correlate a group of subscribed events back to a group of published events by comparing the unique ID of the event block object that a publisher, such as publishing device 1022, attached to the event block object with the event block ID received by the subscriber.
In an operation 916, a determination is made concerning whether or not processing is stopped. If processing is not stopped, processing continues in operation 910 to continue receiving the one or more event streams containing event block objects from the, for example, one or more network devices. If processing is stopped, processing continues in an operation 918. In operation 918, the started projects are stopped. In operation 920, the ESPE is shutdown.
As noted, in some examples, big data is processed for an analytics project after the data is received and stored. In other examples, distributed applications process continuously flowing data in real-time from distributed sources by applying queries to the data before distributing the data to geographically distributed recipients. As noted, an event stream processing engine (ESPE) may continuously apply the queries to the data as it is received and determines which entities receive the processed data. This allows for large amounts of data being received and/or collected in a variety of environments to be processed and distributed in real time. For example, as shown with respect to
Aspects of the present disclosure provide technical solutions to technical problems, such as computing problems that arise when an ESP device fails which results in a complete service interruption and potentially significant data loss. The data loss can be catastrophic when the streamed data is supporting mission critical operations, such as those in support of an ongoing manufacturing or drilling operation. An example of an ESP system achieves a rapid and seamless failover of ESPE running at the plurality of ESP devices without service interruption or data loss, thus significantly improving the reliability of an operational system that relies on the live or real-time processing of the data streams. The event publishing systems, the event subscribing systems, and each ESPE not executing at a failed ESP device are not aware of or effected by the failed ESP device. The ESP system may include thousands of event publishing systems and event subscribing systems. The ESP system keeps the failover logic and awareness within the boundaries of out-messaging network connector and out-messaging network device.
In one example embodiment, a system is provided to support a failover when event stream processing (ESP) event blocks. The system includes, but is not limited to, an out-messaging network device and a computing device. The computing device includes, but is not limited to, one or more processors and one or more computer-readable mediums operably coupled to the one or more processor. The processor is configured to execute an ESP engine (ESPE). The computer-readable medium has instructions stored thereon that, when executed by the processor, cause the computing device to support the failover. An event block object is received from the ESPE that includes a unique identifier. A first status of the computing device as active or standby is determined. When the first status is active, a second status of the computing device as newly active or not newly active is determined. Newly active is determined when the computing device is switched from a standby status to an active status. When the second status is newly active, a last published event block object identifier that uniquely identifies a last published event block object is determined. A next event block object is selected from a non-transitory computer-readable medium accessible by the computing device. The next event block object has an event block object identifier that is greater than the determined last published event block object identifier. The selected next event block object is published to an out-messaging network device. When the second status of the computing device is not newly active, the received event block object is published to the out-messaging network device. When the first status of the computing device is standby, the received event block object is stored in the non-transitory computer-readable medium.
Other examples of the present disclosure may include any number and combination of machine-learning models having any number and combination of characteristics. The machine-learning model(s) can be trained in a supervised, semi-supervised, or unsupervised manner, or any combination of these. The machine-learning model(s) can be implemented using a single computing device or multiple computing devices.
Implementing some examples of the present disclosure at least in part by using machine-learning models can reduce the total number of processing iterations, time, memory, electrical power, or any combination of these consumed by a computing device when analyzing data. For example, a neural network may more readily identify patterns in data than other approaches. This may enable the neural network to analyze the data using fewer processing cycles and less memory than other approaches, while obtaining a similar or greater level of accuracy.
Some machine-learning approaches may be more efficiently and speedily executed and processed with machine-learning specific processors (e.g., not a generic CPU). Such processors may also provide an energy savings when compared to generic CPUs. For example, some of these processors can include a graphical processing unit (GPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), an artificial intelligence (AI) accelerator, a neural computing core, a neural computing engine, a neural processing unit, a purpose-built chip architecture for deep learning, and/or some other machine-learning specific processor that implements a machine learning approach or one or more neural networks using semiconductor (e.g., silicon (Si), gallium arsenide (GaAs)) devices. Furthermore, these processors may also be employed in heterogeneous computing architectures with a number of and a variety of different types of cores, engines, nodes, and/or layers to achieve various energy efficiencies, processing speed improvements, data communication speed improvements, and/or data efficiency targets and improvements throughout various parts of the system when compared to a homogeneous computing architecture that employs CPUs for general purpose computing.
Some embodiments may involve neural network functionality and settings with time series, such as Stacked Model (Neural Network+Time Series) Forecasting Settings (e.g., feature generation settings, model generation settings, model training settings, and model selection options), Neural Network Settings (e.g., task settings, feature generation settings, model generation settings, model training settings, autotune settings, model validation settings, and forecast properties), and Neural Network Model Settings (e.g., dependent variable trend, forecast of dependent variable, dependent variable lags, output layer direct connections, number of hidden layers, number of neurons in layer(s), types of activation functions, random seeds to initialize the network, dependent variable activation function, input standardization methods, dependent variable standardization methods, types of error functions, maximum training iterations).
The stacked modeling strategy can generate forecasts using stacked models that are composed of a neural network model (NN) and a time series model (TS). This modeling strategy can capture the nonlinear relationships between the dependent and independent variables, as well as time series characteristics in the data, such as seasonality and trend. It can model the time series in two steps: in the first step, the neural network model is used to generate the forecasts; and in the second step, the residuals from the first step are passed to the time series model to generate the residual forecasts. In some embodiments, the first and second steps can run sequentially. The final forecasts can be the sum of the forecast from the neural network and the residual forecasts from the time series model.
The pipeline 1100 can be defined in a file, such as a JSON file, indicating the operations to be executed for the pipeline. The file is shareable with other users and formatted to enable the other users to import the pipeline 1100 into a forecasting software environment and execute the pipeline 1100. The pipeline 1100 can be accessed by loading the file of the pipeline 1100 into a software environment and executed by executing the pipeline 1100 using the software environment. The software environment can be a forecasting software environment, which is a software environment configured to enable a user to import and execute pipelines to produce forecasts of time series data. The pipeline 1100 can be implemented by invoking proper services (e.g., microservices) of a distributed computing environment underlying the software environment. This is described in greater detail later in the application with reference to
In the example shown in
The sequence of operations for the pipeline 1100 also include model strategy operations 1104A-C and strategy comparison operation 1106. Each of the model strategy operations 1104A-C is configured to operate on the preprocessed data generated from the preprocessing operation 1102 and to execute, in parallel, modeling strategies 1112A-C involved in the respective model strategy operations 1104 to generate a forecast for each of the multiple time series contained in the preprocessed data. The generated forecasts are analyzed in the strategy comparison operation 1106. Based on the analysis, the champion model strategy for the time-series data 1101 that provided the best prediction results for the time-series data 1101 can be determined. Thereafter, an output operation 1108 can be implemented to determine champion forecasts 1110 for the time-series data 1101 based on the champion model strategy. Each of these operations are further described below in detail.
In some examples, a model strategy used in the model strategy operations 1104A-C involves applying one or more forecasting models to each time series contained in the preprocessed data to produce a forecast for the respective time series. For example, a model strategy 1112A can include an automatic modeling process in which analysis is performed on each time series in the preprocessed data to determine parameters (e.g., optimal or sub-optimal parameters) for one or forecasting models that are to be applied to that particular time series. The one or more forecasting models can then be tuned for and applied to each time series based on the determined parameters, thereby generating a forecast for that time series. Examples of the forecasting models include, but are not limited to, an autoregressive integrated moving average (ARIMA) model, an ARIMA model with exogenous variables (ARIMAX), an unobserved component model (UCM), an exponential smoothing model (ESM), intermittent-demand model (IDM), or any combination of these. Alternatively, some examples can involve model strategies 1112A with predefined configurations or settings (e.g., user-specified customizations or settings) for the one or more forecasting models, which can be applied to each time series to produce a corresponding forecast. In some examples, the model strategy 1112A can specify a single (e.g., comprehensive) forecasting model with which all the time series in the time-series data 1101 are to be modeled (e.g., simultaneously), rather than having a forecasting model specifically tuned for each individual time series.
In some examples, the model strategy 1112A can further involve analyzing the one or more forecasts for each time series to determine a champion forecast for each of the time series. In one example, for each time series, the forecasts generated by the one or more forecasting models in a model strategy are validated to determine a respective error statistic value or error metric. The forecasting model producing the lowest error statistic value is selected for that time series and the lowest error statistic value and the corresponding forecast are associated with that time series in the model strategy.
As one particular example, the model strategy operation 1104A can include three forecast models (e.g., as part of the model strategy 1112A) configured to produce Forecast 1, Forecast 2, and Forecast 3, respectively, for Time Series X in the time-series data 1101. The model strategy 1112A can also be configured to apply the same three forecast models, or a different set of forecast models, to produce Forecast 4, Forecast 5, and Forecast 6, respectively, for Time Series Y in the time-series data 1101. Once these forecasts are generated, the model strategy operation 1104A can be configured to analyze Forecasts 1-3 to determine a champion model for Time Series X and analyze Forecasts 4-6 to determine a champion model for Time Series Y. The analysis for Time Series X can be performed by calculating a performance metric (e.g., a forecast error) for each of the three forecasts and comparing the performance metrics against one another. Examples of the performance metric can include mean absolute percentage error (MAPE), mean absolute error (MAE), mean absolute scaled error (MASE), root-mean-square error (RMSE), or any combination of these. The forecasting model that generated the forecast with the best performance metric, such as the lowest error value, can be selected as the champion forecasting model for Time Series X. The analysis for Time Series Y can be performed similarly. A detailed example of determining a champion model for a time series contained in the time-series data 1101 is described later in the present application with regard to
In some examples, the model strategy 1112A can include user-provided code that is defined in accordance with predefined rules for an application programming interface (API) of the pipeline system. The API can accept certain input parameters and provide corresponding outputs. The input parameters and/or outputs can be provided in any suitable format, such as JSON format. Results of the model strategy can be validated after execution to ensure compliance with the API.
The modeling strategies 1112A-C involved in the model strategy operations 1104A-C of the pipeline 1100 can be saved, retrieved, copied and edited for re-use in different pipelines for the time-series data 1101 or for different time-series data. For example, a user may specify, when generating the pipeline 1100, that a certain model strategy from another pipeline is to be used. In this way, the forecasting models contained in that model strategy as well as the configurations or settings for those forecasting models can be automatically loaded to the pipeline 1100. The user may further edit the model strategy 1112A by, for example, adding or removing forecasting models, changing configurations or settings of the forecasting models, or any combination of these.
The outputs of the different model strategy operations 1104A-C are then merged together and provided as input to the next operation of the pipeline 1100, which in the example shown in
Referring now to
In some examples, Model Strategy 1 employs an automatic modeling process to determine parameters for forecasting models based on each of the multiple time series contained in the time-series data 1101. For instance, Model Strategy 1 can include 15 forecasting models, including both the ARIMA model and the ESM model. The parameters of these 15 forecasting models can be determined based on Time Series 2. These 15 forecasting models can then be tuned with the determined parameters and applied to the Time Series 2 to generate 15 forecasts for Time Series 2 (one forecast from each forecasting model). A performance metric (e.g., a forecast error) can then be calculated for each of the 15 forecasts and compared with one another. Examples of the performance metric can include mean absolute percentage error (MAPE), mean absolute error (MAE), mean absolute scaled error (MASE), root-mean-square error (RMSE), or any combination of these. The forecasting model that generated the forecast with the best performance metric, such as the lowest error value, is selected as the champion forecasting model for Time Series 2. For example, if the ESM model generated a forecast with the best performance metric, the ESM model can thus be selected as the champion forecasting model for Time Series 2 in Model Strategy 1, as shown in Table 1202. This process can be repeated for all of the time series in the time-series data 1101 to produce the Table 1202.
Once the performance metrics for all the champion models in the Table 1202 are determined, a performance metric distribution 1204 (e.g., an error distribution) can be generated for Model Strategy 1. The performance metric distribution 1204 is a distribution of the performance metrics for some or all of the champion models in Model Strategy 1. A similar process can be utilized to generate the Table 1206 and the performance metric distribution 1208 for Model Strategy 2.
The strategy comparison operation 1106 can then compare the performance metric distributions (e.g., performance distributions 1204, 1208) for the various modeling strategies (e.g., Model Strategy 1 vs. Model Strategy 2) to determine the champion model strategy for the time-series data 1101. For example, a combined (e.g., mean, median, or aggregate) statistical value can be calculated for each performance metric distribution and compared to the other combined statistical values associated with the other performance metric distributions to determine the champion model strategy. As one particular example, an aggregate error can be determined for each of the modeling strategies and the model strategy having the smallest aggregate error can be selected as the champion model strategy. Examples of the aggregate error include a weighted mean absolute percentage error (WMAPE), the weighted mean absolute error (WMAE), weighted mean absolute scaled error (WMASE), weighted root-mean-square error (WRMSE), weighted average percent error (WAPE), or any combination of these.
Referring back to
While the pipeline 1100 in
The segmented pipeline can also be defined in a file, such as a JSON file, with code indicating the sequence of operations to be executed for the segmented pipeline.
The sequence of operations included in the segmented pipeline 1300 further includes a segment operation 1304 involving splitting the time-series data up into data segments (groups) that can be operated on in parallel. Each data segment has a common characteristic. For example, the segment operation 1304 can split up the time-series data into data segments by type, source, quantity, associated entities, or any combination of these. In some examples, the segment operation 1304 involves analyzing metadata (e.g., an indicator or a flag) associated with each of the time series in the time-series data to determine an appropriate data segment for the time series. In another example, the segment operation 1304 involves determining the data segments using other mechanisms, such as based on the analysis of the nature of the time series contained in the time-series data.
Each of the data segments can be assigned to its own pipeline segment (e.g., branch) and operated on in parallel to the data segments. For example, each of the pipeline segments 1310 corresponds to one of the data segments and can include a group of operations. The group of operations can include a data collection and processing operation 1306, one or more model strategy operations 1308, and a strategy comparison operation 1310.
The data collection and processing operations 1306a-d can each involve collecting data for a data segment that corresponds to the pipeline segment 1310. The data collection and processing operations 1306a-d can also each involve preparing the data segment so that it can be processed by the corresponding pipeline segment 1310. The data collection and processing operations 1306a-d can additionally or alternatively involve other processing operations, such as processing the corresponding data segment to make them suitable for the modeling strategies contained in a pipeline segment, or further dividing the data segments into groups so that different forecasting models in the pipeline segment can be properly applied.
The model strategy operations 1306 and the strategy comparison operations 1310 contained in the pipeline segments 1310 can be implemented similarly to the model strategy operation 1104 and the strategy comparison operation 1106, respectively, as described above. One of the differences is that a model strategy operation 1306 applies the model strategy to its corresponding data segment, rather than the entire input time-series data, to generate forecasts. Likewise, the strategy comparison module 1310 analyzes performance metrics associated with the forecasts and determine a champion model strategy for the corresponding data segment.
The sequence of operations in the segmented pipeline 1300 further includes a merge segment operation 1312, in which the forecasts generated by the pipeline segments 1310 are merged together to generate the final forecast for the input time-series data. The merging can include interleaving, concatenation, or both. The merge segment operation 1312 may further place the forecasts in a format that is the same as the input time-series data or as specified by the user requesting the forecast. The sequence of operations can also include an output operation 1314, which can be similar to the output operation 1108 described with regard to
The user interface shown in
Adding a model strategy to the pipeline 1100 can be performed by dragging and dropping the icon corresponding to the selected model strategy from the panel 1422 to a proper position of the graphical visualization 1408, such as after the data operation node 1410. Upon dropping the selected model strategy at the proper position, an operation node corresponding to the model strategy can be automatically added to the graphical visualization 1408 and appropriate connections (e.g., a connection from the data operation node 1410 to the operation node and a connection from the operation node to the model comparison operation node 1418) are also automatically added to the graphical visualization 1408. The pipeline 1100 can be updated accordingly to include the newly added operation and the connections.
If the newly added model strategy is the first model strategy in the pipeline, the output operation node 1420 is also added automatically along with the model strategy operation node corresponding to the newly added model strategy. If a second model strategy is added in the pipeline, a model comparison operation node 1418 and any connections between the model comparison operation node 1418 and other nodes in the pipeline are automatically added along with a model strategy operation node for the second model strategy. Likewise, if an existing model strategy operation node is removed, any connections associated with the model strategy operation node are also removed. If the removed model strategy operation node is the last model strategy operation node in the pipeline 1100, the model comparison operation node 1418, the output operation node 1420 and any connections between them are also automatically removed. The pipeline 1100 is also updated accordingly, such as by modifying the JSON file describing the pipeline 1100 based on the changes to the graphical visualization 1408.
Although the above example focuses on a model comparison operation node 1418, the automatic addition or deletion of related nodes can also be applied to other types of nodes. When a user adds an operation into the pipeline, other operations may also be automatically added to (or removed from) the pipeline according to predefined rules. The predefined rules can define required operations for a given operation. For example, the predefined rules can define an output operation as a required operation for a model strategy operation and a data operation. As another example, the predefined rules can define a model comparison operation as a required operation for multiple model strategy operations, but not for a single model strategy operation. Based on these rules, required operations can be automatically added to the pipeline when a particular operation is added by the user. The graphical visualization of the pipeline may also be automatically updated to include visual representations of these additional required operations, such as the nodes shown in the GUI 1400, along with any suitable connections between adjacent operations in the pipeline.
Likewise, various operations that become extraneous when an existing operation is removed from the pipeline. So, the extraneous operations can be automatically removed from the pipeline, in accordance with the predefined rules. The graphical visualization of the pipeline may also be automatically updated to remove visual representations of the existing operation and any other extraneous operations.
The GUI 1400 further shows a group of control buttons 1424 that can be utilized to control the pipeline 1100 represented by the graphical visualization 1408. For instance, a user can save or delete the pipe 1100 by clicking on or otherwise activate the “save” icon 1430 or the “delete” icon 1436, respectively. The user can also execute or stop the execution of the pipeline by clicking on or otherwise activate the “play” icon 1432 or the “stop” icon 1434, respectively.
The segmented pipeline 1300 can be executed by activating the “play” icon 1532. The status of the execution of each operation can be shown via the corresponding operation node. In the example GUI shown in
Further, as shown in
A champion pipeline can be determined from among the multiple pipelines in a project, e.g., by determining the pipeline having the smallest aggregate error. A marker or other visual indicator can then be depicted, in the “Pipeline Comparison” user interface 1600, proximate to the champion pipeline to indicate that it is the champion pipeline. Similarly, multiple projects can also be established and compared. For example, multiple projects can be compared by comparing the pipelines within the projects. A champion pipeline can be determined for each of the multiple projects. The champion pipelines for these projects can be compared with one another to determine the champion project, which may be the project containing the champion pipeline having the smallest aggregate error.
Although the above description focuses on a user manually creating a pipeline in the GUI 1400 or 1500, a pipeline or a portion there of may be automatically generated based on a rule set. For example, a user can specify high level parameters that the user wants in the pipeline, such as the number of segments, the number of model strategies, the default model strategies to be included in the pipeline, or a combination thereof. Based on the ruleset, the computing device can automatically generate the pipeline and its association graphical visualization.
Referring now to
In block 1702, a processing device obtains time-series data for forecast. The time-series data can include multiple time series. The time-series data can be provided by a client device or be collected by the processing device from one or more third-party devices, such as being downloaded from one or more cloud computing servers or from a distributed network configured for measuring or otherwise collecting the time-series data.
In block 1704, the processing device obtains one or more pipelines describing a sequence of operations for processing the time-series data to produce one or more forecasts. In one example, the pipelines are generated based on inputs sent from a client device. For example, the processing device can generate a GUI, such as the GUIs 1500-1600, and have the GUI presented on the client device. A user of the client device can operate in the GUI to build the pipelines, execute the pipelines, view the forecasts or perform any combination of these operations. In another example, the processing device can obtain the pipelines by accessing files defining existing pipelines that are used for other time-series data and importing the files into the current forecasting software environment.
In block 1706, the processing device accesses one of the pipelines for evaluation. This pipeline is referred to below as the “current pipeline,” since this is the pipeline to be currently evaluated and executed.
In block 1708, the processing device determines whether the current pipeline is a segmented pipeline. As discussed above, a segmented pipeline includes multiple pipeline segments (e.g., branches) each configured to perform the forecasting for a particular segment of the time-series data.
If the current pipeline is not a segmented pipeline, in block 1709, the processing device performs the preprocessing operation of the pipeline to preprocess the time-series data, such as to clean, add, remove, or re-format the time-series data.
In block 1710, the processing device executes the pipeline. This may involve executing one or more model strategy operations on the preprocessed data as defined in the pipeline. The processing device may compare the results of the model strategy operations to determine a champion model strategy and a champion forecast. Additional examples of the processing device executing the pipeline are described below with regard to
If the current pipeline is a segmented pipeline, the process 1700 proceeds to block 1712. In block 1712, the processing device preprocesses the time-series data, for example, as described above in relation to block 1709. In addition, the processing device divides the time-series data into data segments that correspond to the pipeline's segments.
In block 1714, the processing device executes each of the pipeline segments by applying the operation(s) in each respective pipeline segment to the corresponding data segment, such as by applying modeling strategies on the data segment and comparing the results of the modeling strategies. In some examples, executing the pipeline segments may involve implementing the process described later on in this disclosure with regard to
In block 1716, the processing device combines the results of the pipeline segments to generate a champion forecast for the segmented pipeline.
In block 1718, the processing device determines whether there are more pipelines for the time-series data. If so, the processing device accesses the next pipeline for evaluation and repeats steps 1708-1718 for the next pipeline.
Once there are no more pipelines to be evaluated, the process 1700 proceeds to block 1720. In block 1720, the processing device determines a champion pipeline (assuming that there are multiple pipelines for the time-series data). As discussed in detail above, the comparison can be performed by comparing distributions of performance metrics between the pipelines. Based on the distribution comparison, the processing device can select a champion pipeline.
In some examples, the processing device may further receive an override instruction from a user to instead use another pipeline as the champion pipeline, rather than the system-designated champion pipeline. Subsequent to receiving the override instruction, the processing device can specify the other pipeline as the champion pipeline for the time-series data.
In block 1722, the processing device generates forecasts for the time-series data using the champion pipeline determined in block 1720.
It should be understood that while the above describes that the pipelines are processed one after another, the pipelines can be processed in parallel in a distributed computing environment. Furthermore, while the above describes that a processing device performs the various operations involved in the process 1700, the phrase “a processing device” as used herein is intended to encompass one processing device or multiple processing devices. Any number and combination of processing devices can work in concert to implement the examples described herein.
In block 1801, a processing device accesses input time-series data. If the process 1800 is performed for a pipeline segment, the input time-series data contains the data segment that is generated for this pipeline segment, such as the data segment generated in block 1712 of
In block 1802, the processing device accesses the next model strategy operation contained in the pipeline or pipeline segment.
In block 1804, the processing device executes the model strategy operation, which may include one or more model strategies. One exemplary model strategy can include an automatic modeling process in which the input time-series data are analyzed to generate forecasting models for the time-series data. Another exemplary model strategy can include a single comprehensive forecasting model in which all the time series in the time-series data are modeled simultaneously. Another exemplary model strategy can include user-provided code rather than using existing forecasting models. The user-provided code may be defined in accordance with an existing API or “contract.” In some examples, the user-provided code may include forecast values explicitly provided by the user, in which case no additional forecasting need to be performed for the time-series data. Execution of the model strategy operation can involve generating at least one forecast for each of the time series contained in the input time-series data.
In block 1806, the processing device determines the performance metrics of the forecast results generated by the model strategy, such as the forecast errors measured by MAPE, MAE, MASE, or RMSE.
In block 1808, the processing device determines whether there are more model strategy operations to be performed for the pipeline or pipeline segment. If so, the next model strategy operation is accessed in block 1802 and steps 1084-1808 are repeated for the next model strategy operation.
Once all the modeling strategies for the pipeline or pipeline segments have been executed, the process 1800 can proceed to block 1810. In block 1810, the processing device executes a strategy comparison operation to determine a champion model strategy and/or a champion forecast for the time-series data. The comparison is performed by comparing the distributions of performance metrics for the modeling strategies, such as by calculating and comparing aggregate error values based on the error distributions of the modeling strategies.
In block 1812, the processing device outputs the forecast generated by the champion model strategy as the champion forecast for the input time-series data.
In order to manage or coordinate the operations, the pipeline service 1906 can invoke the operation service 1908 for each operation of the pipeline. The operation service 1908 load the parameters and code for the corresponding operation and invoke other microservices that can be utilized to execute the operation, such as the compute service 1910 and the comparison service 1912. The compute service 1910 can execute the code implementing the corresponding operation in the forecasting software environment to generate output results, such as executing the model strategy operations. After the computing service 1912 finishes the respective execution, the operation service 1908 further invokes the comparison service 1912 to compare the sets of the forecasts generated. The comparison service 1912 can be invoked to determine the champion model strategy or the champion pipeline. The various microservices described above can be executed in a distributed computing environment 1914. While
In the previous description, for the purposes of explanation, specific details are set forth in order to provide a thorough understanding of examples of the technology. But various examples can be practiced without these specific details. The figures and description are not intended to be restrictive.
The previous description provides examples that are not intended to limit the scope, applicability, or configuration of the disclosure. Rather, the previous description of the examples provides those skilled in the art with an enabling description for implementing an example. Various changes may be made in the function and arrangement of elements without departing from the spirit and scope of the technology as set forth in the appended claims.
Specific details are given in the previous description to provide a thorough understanding of the examples. But the examples may be practiced without these specific details. For example, circuits, systems, networks, processes, and other components can be shown as components in block diagram form to prevent obscuring the examples in unnecessary detail. In other examples, well-known circuits, processes, algorithms, structures, and techniques may be shown without unnecessary detail in order to avoid obscuring the examples.
Also, individual examples may have been described as a process that is depicted as a flowchart, a flow diagram, a data flow diagram, a structure diagram, or a block diagram. Although a flowchart can describe the operations as a sequential process, many of the operations can be performed in parallel or concurrently. In addition, the order of the operations can be re-arranged. And a process can have more or fewer operations than are depicted in a figure. A process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc. When a process corresponds to a function, its termination can correspond to a return of the function to the calling function or the main function.
Systems depicted in some of the figures can be provided in various configurations. In some examples, the systems can be configured as a distributed system where one or more components of the system are distributed across one or more networks in a cloud computing system.
| Number | Date | Country | Kind |
|---|---|---|---|
| 201811026489 | Jul 2018 | IN | national |
This claims the benefit of priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application No. 62/866,595, filed Jun. 25, 2019, U.S. Provisional Patent Application No. 62/700,966, filed Jul. 20, 2018, and to U.S. Provisional Patent Application No. 62/690,169, filed Jun. 26, 2018, and the benefit of priority under 35 U.S.C. 119(b) to Indian Provisional Patent Application No. 201811026489, filed Jul. 16, 2018, the entirety of each of which is hereby incorporated by reference herein.
| Number | Date | Country | |
|---|---|---|---|
| 62866595 | Jun 2019 | US | |
| 62700966 | Jul 2018 | US | |
| 62690169 | Jun 2018 | US |
